Conventional techniques that are employed to recommend content to consumers typically employ an “all-in-one” approach that leaves little room for the consumer to customize how the system generates the recommendations. These conventional techniques also typically fail to consider that the user may have different media consumption modalities. One result of these shortcomings is that the consumer often discounts the recommendations made using these conventional techniques, limiting the intended effect and utility of the techniques.